energy prices. industrial struc'tore and choice of
TRANSCRIPT
No. 79
A list of Horking Papers on
the last pages
ENERGY PRICES. INDUSTRIAL STRUC'TORE
AND CHOICE OF TECHHOLOGY: An Inter
nationa1 Comparison with Specia1
Empbasis on the Cement Industry
by
Bo Carlsson
This is a preliminary paper. It is intended for private circulation, and should not be quoted or referred to in publications without permission of the authors. Comments are welcome.
February 1983
Energy Prices, Industrial Structure and Choice of Technology: An International Comparison with Special Emphasis on the Cement Industry
by Bo Carlsson The Industrial Institute for Economic and Social Research, Stockholm
L Introduction
The economic and industrial structure of a country is determined by a number of factors. Among these are natura1 resources, the qua1ity and quantity of 1abor, historical traditions, etc. In order to understand the economic development of a country it is, therefore, necessary to adopt a long-term perspective. For example, it of ten takes a couple of decades for new technologies to be developed and yet a couple of decades for them to replace older technologies. Thus, today-s industrial structure is determined to a large extent by prices and other factors during the 1950-s and 1960-s or even further back.
In a recent study for the Swedish Energy Commission,l) I have studied the development of energy prices in relation to other prices in Sweden, Great Britain, West Germany, and the United States during most of the 20th century. Some of the results of this comparison are illustrated in section II below. The relative price differences are also shown to be reflected in the energy consumption in various industries in Sweden, West Germany, and the United States. The energy consumption varies due to differences in output mix and choice of technology.
Section III and the following try to explain why the choice of technology varies international ly. Obviously, there are many factors besides relative prices which playarole, but in order to provide a more complete framework it is necessary to carry out the ana1ysis at a quite detailed levpl.
1) B .. Car~sson, "Relativprisutvecklingen på energi och dess betydelse för energlåtgang, branschstruktur och teknologival i en internationell jämförelse" Crhe Oe~e10pment of Relative Energy Prices and Its Impact on Energy Use, Indust~la1 Structure and Choice of Technology: An International Comparison") App~ndlx 12 to the report to the Energy Commission by the Expert Group on P011CY Instruments. OS I 1977:17, Stockholm 1977. Also published by the Industrial Institute for Economic and Social Research, Stockholm. A more complete version of the cement industry study has been published (in Englis~): B. Carlsson, "Choice of Technology in the Cement Industry: A Comparison of the United States and Sweden", in B. Carlsson, G. Eliasson and I. Nadiri (eds), The Importance of Technology and the Permanence of Structure in Industrial Growth, The Industrial Institute for Economic and Social Research, Stockholm, 1978.
For severa1 reasons the cement manufacturing process has been chosen for this study: Cement manufacturing is one of the most energy consurning processes in the whole of manufacturing industry; the output is homogeneous; the production process is relatively uncomp1icated and separable from other processes; and it is known from the start that the choice of production techniques has been very different in various countries, at 1east up unti1 recently.
Section III describes the cement manufacturing process and provides a
2.
brief history of the technological development of the industry. Section IV brings out some salient features of the industry and how they differ between Sweden and the United States. This analysis is based largely on interviews conducted by the author during the spring of 1977 in both the United States and Sweden. In Section V, some investment cost calculations for both wet and dry kilns using price data for 1970 and 1975 will be presented. Section Vldiscusses the differences between actual and theoretical costs of wet and dry process plants and section Vllanalyzes the reasons for the delayed introduction of the suspension preheater process. Section ~III concludes the study.
II. Relative Prices and Energy Consumption in Manufacturing Industries
Although energy prices have generally kept pace with other industrial prices throughout the postwar period unti1 1973-74, they have fallen quite drastically in relation to industrial wages. This is shown for West Germany, Sweden and the United States in figure 1. Prices for various kinds of energy have been weighted together to an average energy price for industrial consumers in each country. The figure shows how many working hours are equiva1ent in cost to one MWh (= 1000 kWh). This relative price of energy has been much lower in the United States than in Sweden and West Germany throughout the whole period 1930-1975, but the international differences have diminished over time.
As shown in figure 2, energy prices have generally been lower in the United States than in Western Europe not only in relative but also in absolute terms. The figure shows the price of energy in Swedish currency, all energy having been converted to kWh. Average industrial energy prices were 20 percent higher in West Germany than in Sweden in 1965, whi1e those in the United States were about 50 percent lower than the Swedish ones. However,
3.
Swedish electricity prices have been the lowest "among the three countries, due large1y to the avai1ability (at least until 1960) of cheap hydra power.
Based on this development of relative energy prices, one would expect the following to be true:
l) Of the three countries, the United States ought to have the largest energy consumption per unit of output, since both absolute and relative energy prices have been the lowest there. For the opposite reason, West Germany ought to have the lowest energy consumption per unit of output.
2) Since electricity prices have been lowest in Sweden in relation to fuel prices, one would expect alarger share of electricity in industrial energy consumption in Sweden than in the other countries.
3) Since electricity prices have been lower in Sweden relative to wages than in the rest of Western Europe, one would expect alarger share of processes heavily dependent on electricity in Sweden than in other countries.
An examination of table l will show that these hypotheses are largely confirmed. The United States turns out to have the highest total energy consumption in most industries, West Germany the lowest, with Sweden somewhere in between. However, for manufacturing industry as a whole, Sweden is shown to have the highest total energy consumption per dollar of value added, due to a more energy consuming output mix.
As regards electricity use, Sweden turns out to have somewhat higher consumption per unit of output than the other countries. This may indicate certain possibilities of substitution between electricity and other types of energy. But the main reason why Sweden has about twice as much electricity use per unit of output in total manufacturing as the United States and West Germany is that Sweden has extremely high electricity consumption in four industries. These are the paper and pulp industry, the chemical industry, the iron and steel industry, and the non-ferrous metals industry. In these industries, the high electricity consumption is due to a large share of certain processes: mechanical pulp, electro-chemical processes, e1ectric steel furnaces, and aluminium smelting. Thus, there are indications that the relatively low electricity prices in Sweden have contributed to specialization on certain types of electricity using production.
We turn now to an analysis of why the choice of production process varies among countries.
9.00
8.00
7.00
6.00
5.00
4.00
3.00
2.00
1.00
Figure 1
Working ours/MWh
1930 40
Relative Prices of Energy and Labor in Manufacturing in Sweden, West Germany and the United States, 1930-1975.
50 60
. . . . . . . . . ,,' ..... "
70 75
West Germany
Sweden
Figure 2
örejkWh
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
Energy Prices in Manufacturing Industry in Sweden, West Germany and the United States, 1930-1975. Öre/kWh (current prices).
.. .. ' ,,' " .
.' . ' •• , t'..
· · · · · · · · · · · · · · · . . . .
E1ectricity (West German~
E1ectricity (USA)
E1ectricity (Sweden)
All energy (West Germany)
All energy (Sweden)
All energy (USA) 2-L_..---~ ,/ ,-,-
... _-----_ ... _----_ ........... ' ---_. __ ._-----~ .. ~ ... ~ Year
1930 40 50 60 70 75
Sources: See B. Carlsson, "Relativprisutvecklingen på energi ... ", op. dt.
5.
Table 1 Energy Consumption per Unit of Output in Certain Manufaeturing Industries in the United States, Sweden and West Germany 1967
kWh per dollar of value added.
Industry
Food, beverages and tobaeeo Textiles, apparel, leather
and leather goods Lumber, wood produets and
furniture Pulp and paper Mise. eonverted paper and
paperboard produets Printing and publishing Industrial ehemieals, plasties
materials and syntheties, agrieultural ehemiea1s
Other ehemiea1 produets Petroleum refining Paving and roofing materials,
mise. petroleum and eoal produets
Rubber produets Plasties produets Stone, e1ay and glass produets Blast furnaee and basie steel
produets, iron and stee1 foundries and forgings
Nonferrous meta1s Fabrieated meta1 produets Maehinery, exeept eleetriea1 Eleetrieal equipment and
supplies Ship and boat building and
repairing Other transport equipment Instruments and related
produets
Mise. manufaeturing industries
Total manufaeturing industry
Sourees: SOS Industri 1968.
Total Energy Consumption Thereof E1eetrieity West - West
USA Sweden Germany USA Sweden Germany
9,4
5,6
7,5 71,0
7,3 1 ,3
8,9
5,6
4,5 72,4
3,4 l ,4
4,4 1,0
4,8 1,2
2,1 1,1 56,0 10,2
4,8 l , l
l ,4 0,4
1 ,2
0,9
1 ,5
21 ,1
0,8 0,4
0,4
0,8
0,5 8,7
0,8 0,4
52,3126,633'~8,6 20,9 6,~ 4,9'
8'~,218,7)9,6 4,9 o,~ o,~
81,5 23,2 21,1 4,6 3,9 1,6
32,2 9,4 6,0
43,2
38,3 29,9 4,8
3,3
2,8
3,3
3,8
2,4
2,8
13,4
6,3 9,6 5,2
35,4
55,0 27,4 4,7 4,2
2,7
4,2 4,4
1,4 2,1
15,0
41,5 1,0 7, l 1 ,6
4,6 l ,6
28,5 2,5
54,6 4,5
17,8 10,8 3,9 0,8 2,3 0,6
2,0 0,8
3,1 0,8 4,3 0,8
1 ,3 0,5 l ,3 0,5
11,6 2,0
0,8 1 ,8
l ,2
3,0
10,7 14,9
1 ,3
1,0
0,9
1 ,0
1 ,0
0,4 0,5
3,7
2,4
1 ,3
l ,2
2,2
3,9 5,4 0,7
0,4
0,5
0,9 1 ,0
0,2 0,3
l ,6
U.S. Bureau of the Census, 1967 Census of Manufactures, Vol. II, Industry Statistics, Part l, Major Groups 20-24, U.S. Government Printing Office, Washington, D.C., 1971. Census of Manufacturers, 1967. Special Series: Fuels and Electric Energy Consumed, MC 67(s)-4, U.S.G.P.O., Washington, D.C. 1971, table 4. Statistisches Jahrbuch fur die Bundesrepublik Deutschland 1969 Statistisches Bundesamt, Wiesbaden, 1969, table WII.9 and XII.lO .,...., . .,
6.
III. The Cement Manufacturing Process - Description and Brief History
The raw material for cement production consists mainly of limestone which is crushed and then ground into a fine powder. In the dry cement manufacturi ng process, the powder i s fed di rectly i nto a kil n where it is calcined (burned) to form clinker. In the wet process, water is added to form a slurry which is then fed into the kiln. The ki1n is essentia1ly a huge cylindrical steel rotary tube lined with firebrick. Some kilns have a diameter up to 8 meters and are up to 230 meters long -- longer than
the height of a 70-story building. The kiln axis is slightly inclined, and the raw material (efther slurry.or dry)· is fed into the upper end. At the lower end is an intensely hot flame which provides a temperature zone of about 15000 C by the precisely controlled burning of coal, oil or natural gas under forced draft conditions. l ) Af ter the c1inker is cooled, it is ground with 4-6 % gypsum into cement.
The earliest cement kilns were dry process but of a different type (vertical shaft kilns) than the modern ones. In the early 1900's, long rotary horizontal kilns began to be introduced. Because of the relative ease of grinding and homogenizing the raw materials under wet conditions, the wet process came to dominate. The drawback of the method, however, is that it is much more fuel consuming than the dry process, since the water added in the raw mill must be dried away before calcination can take place.
In 1927, a semi-dry process was patented in Germany. It was named the Lepol process (acronym for the inventor, Lellep and the equipment manufacturer, Polysius).2) The basic principle of the process is to use the exhaust gases from the kiln for drying and preheating the raw materials before inserting them into the kiln, Thus, the main advantage is energy saving. The process became popular in Europe but was hard1y used at all in the United States.
1) Energy Conservation Potential in the Cement Industry, Conservation
Paper number 26, prepared by the Portland Cement Association for the Federal Energy Administration. June 1975 (Springfield, Va.: U.S. Department of Commerce, National Technical Information Service, PB-245 159), p. 1.
fvl):ingel, Technischer Fortschritt, Wachstum und Konzentration in der
~tindustrie, doctora1 dissertation, 1972, p. 24~
7.
In 1933, yet a new type of dry process cement kiln was patented in
Czechoslovakia. Then Wor19 War II intervened, but af ter the war the patent was acquired by a German equipment manufacturer, and the first installation was made in 1950 in Germany.l) In a conventional dry kiln, three sub-processes take place simultaneously. At the upper end, where the materials are fed into the kiln, preheating takes place. In the middle, calcining
-occurs, and at the lower end the final burning. Since only a fraction of the raw materials on the rotating kiln wall is exposed to the hot air, the heat exchange is very inefficient and the fuel consumption therefore high. Also, since the sub-processes require different temperatures, it is difficult to optimize the temperature throughout the kiln, and different scale factors seem to apply. A number of interviews conducted by the author have indicated that difficult operational problems arise in long conventional kilns as the scale is expanded.
The essence of the new kiln is that it separates the preheating from the other sub-processes which take place in a conventional kiln. Preheating of
the materials takes place in cyclones where hot air from the kiln is blown in the opposite direction to that of the powder, with the result that the powder is temporarily suspended in the air. This provides a much more efficient heat exchange between the air and the materials than can be achieved in a kiln and the amount of energy required is therefore reduced very significantly.
In recent years, Japanese firms with license rights on West German suspension preheaters have developed an auxiliary burner system in the preheater, so that not only preheating but also up to 95% of the ca1-cina~ion takes place before the feed en ters the kiln. In such a precalcining system both the length and the diameter of th~ kiln can be further reduced, and energy consumption may also be slightly reduced. But the main advantage of the precalcing system may lie in its ability to
deal wit~ some operational problems encountered in suspension preheater
l) Hoke M., Garrett, "The Potential Promise - Prospects and Pitfalls in Energy Conservation by the U.S. Cement Industryll, in Proceedings of the, FEA-PCA Seminar on_Energy Management in the Cement Industry, Federal Energy
Administration Conservation Paper Number 47, FEA/D-76/09l, p. 268.
8.
Figure 3 Technical Change in Cement Kilns
1.
2.
3.
4.
r;;::s::j Drying f;;:··;·;·;·;.;J Preheating
c::::J Calcining 5. ~ Burning lZZ3 Cooling
L Conventional long (dry) kiln.
2. Ory k il n with l-stage preheater.
3. Ory ki 1 n with 2-stage preheater.
4. Ory ki l n with 4-stage preheater
5. Ory kiln with 4-stage preheater and precalciner.
Source: F.L. Smidth & Co.
9.
systems1). The first precalcining system was developed in Japan in 19662). The process has already been introduced in the United States (1976) and is currently being introduced in Sweden.
The development of cement production technology over the past 30 years is illustrated in figure 1. Until 1950. conventional long kilns were used. With the arrival of cyclone preheaters, the length and diameter of the kiln could be substantially reduced. In order to produce l 225 tons in 1950, d kiln of 143 meters and 4.8 meters~ diameter was required. In the 1970~s,. a kiln of 63 meters and a diameter of 4.2 meters could
produce the same output. 3)
With the preheating of the materials taking place outside the kiln, the
length and the diameter of the kiln can be substantially reduced for the same capacity. This, in turn, means a (theoretical) saving in capital cost, since
preheater cyclones are cheaper to build and install than the additional kiln section which would otherwise be required. Alternatively, for the same capital cost, much larger capacity can be obtained. Since the number of people required to operate the kiln is about the same, no matter what size
and type of the kiln, the suspension preheater process also offers substantial labor saving.
IV. Industry Comparison
IV.l Industry Characteristics . ~ - - - - ~ - - ~ - --There are four characteristics of the cement industry which go a long way towards explaining the differences between the Swedish and the American cement industries observed above. These are economies of scale, high energy intensity, high transport costs, and homogeneous output.
f) G d" . or 1an Assoc1ates, Industr1al International Data Base, The Cement Industry, NATO/CCMS-46. New York: Energy Research and Development Administration. 1976, p. 14.
2) FEA-PCA Proceedings, p. 267.
3) H.R. Norbom, "Wet or Dry Process Kiln for your New Rock Products. Vol. 77. No. 5 (Mav. Installation?," 1974). PP. 92-93.
10.
a) Economies of Scale
The presenee of economies of scale in the cement industry is illustrated
in figure 4. There are substantial economies of scale in both wet and dry plants. The investment cost per ton of annual capacity is lower (at least theoretically) for dry than for wet plants and continues to decrease beyond
where the investment cost per ton in wet plants levels off.
b} High Energy Intensity
The cement industry is extremely energy intensive. In Sweden, the fuel and electricity east has ranged between 29 and 41 % of the value of sales during the period 1950-75. In the United States, the corresponding range was 19 to 28 %. The energy eos t has been higher than the labor cost throughout the period studied in both countries. l We will return later to the energy considerations in detail when discussing the choice of
technology.
c) High Transport Costs
Because of the relatively low price per ton, the relative
transport costs of cement are extremely high. This means that the cement industry is highly 10ca1 in character, especially in regions without water transport facilities. It costs as much to transport a ton of cement 100 km by truck in Sweden as 600 km by small coastal shipping vessels or 2 000 km by large bulk carries. 2 Therefore, in order to utilize scale economies fully, cement plants must be located either in large metropolitan areas or on v/aterv/ays.
Because of the high transport costs not only for the finished product but a1so, and even more so, for raw materials, the cement industry is forced to rely on 10ca1 raw materials which may vary greatly in quality among locations. Thus, the moisture content and purity of the raw materials as well as their hardness and accessi.bility vary substantially among plants.
d} Homogeneous Output
Although the quality of cement can theoretically vary among plants and 10cations, most countries impose fairly stringent requirements which must
l} See figures 7 and 8 be1ow.
2) B. Carlsson, "Industrins energiförbrukning 1974-80" (Industria1 Energy Consumption 1974-80"), Appendix 7 to IUI:s lånotidsbedömning 1976 (IUI~s Medium Term Survey 1976), IUI, Stockholm: 1977, p. 277.
, 11.
Fi gure 4.. Investment Costs ; n Wet and Dry Cement P1 ants, 1970 and 1975
~ i ~ r 30 l ~ I so! B 120
6J I ~ 'O I o.. 4 r
---- Wet plants
---- Dry plants
I 0,2 0,4 0,6 ~.8 1.0 Capacity, million annual tons o o L-__ ~,~ ____ ~, __ ~-,, __ ~~ __ -~--~ ____ ~~ __ ~,
500 :000 1500 2000 2~O~ JOCJ J500 &000
Capacity? tons per day
Sources: K.T. Andersen, "Kiln Selection ", in Proceedings of the FEA-PCA Seminaron Energy Management in the Cement IndUStry, Conservation paper No. 47,~75, p. 207. S. Mänqel, Technischer Fortschritt Wachstun und Konzentration in der Deutschen Zementindustrie, doctoral dissertation, pp. 47-48.
12.
be met by cement sold domestically. These requirements pertain to compressive strength, fineness, chemical composition, etc. They vary somewhat among countries, although the differences are not great among Western countries. It does seem, however, as though the U.S. specifications are
more stringent in terms of both fineness and purity (esp. concern-ing the presence of alkalis) than those of West European countries~) The fact that U.S. cement is more finely ground essentially leads to slightly
higher energy consumption than would otherwise be the case.
The stricteralkali requirements may have more far-reaching consequences for the choice of technology, however, as will be shown below.
The presence of substantial economies of scale in combination with high transport costs has imDortant implications for market structure. In Sweden, six out of a total of seven cement plants are located near water. This has made it possible to take advantage of scale economies in produc
tion. Thus, the average Swedish cement plant had a capacity in 1975 of 725 000 tons, while the average American plant had a capacity in 1976 of only 545 000 tons. 2) In addition, because of an extremely high degree of concentration (there is now only one domestic cement firm in Sweden af ter a merger in 1974), it has been possible to plan the production expansion in such away that there is very little overlap geographica11y between plants. The primary reason why the Swedish government a110wed the merger to go throus'
was precisely the argument that this would facilitate achieving a more optimal industry plant structure, provided at the same time that there would be no tariff or other protection, and that the company would be subjected to price control. The product on capacity of the Swedish cement company is large enough to place it among the four largest U.S. firms.
By contrast, there were 52 cement companies in the United States in 1976, the largest of which had 6.7 % of industry capacity. The four and eight largest firms accounted for 22.3 and 39.2 % respectively.3)
1) Gordian Associates, op. cit., p. 39.
2) Portland Cement Association, op. cit., and Cementa AB.
3) Portland Cement Association, op.cit., p. 3
13.
The plants within the largest firms are also widely scattered geographica11y,making it difficu1t to concentrate production to one 10cation without involving major changes in regional market shares. There were 162 plants in the U.S. in 1976. This large number can be explained by both geographical factors (population density, transport costs, large inland areas without access to water transport facilities, etc.) and historical factors (most plants were built when scale advantages were less pronounced in areas where cement was needed and local raw materials were available).
While the above factors explain the differences in plant structure between the U.S. and the Swedish cement industries, these are also differences in the size structure of kilns. In 1975, the average Swedish cement kiln had an annual capacity of 250 000 tons, whi1e the corresponding figure for the U.S. was 212 000 tons. At the same time, Swedish dry kilns were 50 % larger than U.S. dry kilns. l )
The reasons for these differences are related to those explaing the in plant structure. During the last fifteen years, kilns
built in the United States have tended to be relatively small. Immediately af ter the Second World War there was a shortage of cement capacity in the United States which led to overinvestment in the 1950~s and early 1960~s. The resulting overeapacity seems to have lasted into the early 1970~s, making it unattractive to invest in anything but replacements of old, inefficient facilities. Since replacing an old wet kiln by a suspension preheater system would involve replacing much of the raw material handling equipment as \'/ell, there is a certain minimum scale below which
the capital cost would be prohibitive.
Labor Productivity
. ,
Even though both kilns and plants tend to be larger in Sweden than in the United States, labor productivity in the United States has remained higher than in Sweden throughout the period. See figure 5. However, the
labor productivity gap has narrowed from 49 % difference in 1950 to 25 % in 1974. On the other hand, it is also shown in figure 5 that the total wage cost per hour has increased far more rapidly in Sweden than in the U.S., so that in 1975 the Swedish .w~ge rate exceeded the American one.
l) Portland Cement Association, op.cit. and Cementa AB.
FigurE~ Deve10pment of wage rates and labor productivity in the cement
industry 1950-75 in the United States and Sweden
14.
Labor Input Man-hours per l 000
metric tons of cement
Wage rate
$/hour
, ..... ,
2 500
2 000
500
l 000
500
, ""
Labor Input
" , , "" ,
" , , ,
ISweden I I I I USA I
", 'J " I Hage ra te
, I , I
' ...... "'"' I " I '\ I
\ I
" I '-.-/ /'0./ ...... / .......... ~
/ " Sweden /
United State s
8
7
6
5
4
3
2
1
Year 1950 55 60 65 70 75
Note: U.S. figures inc1ude both direct and overhead labor. The Swedish figures have been made comparable in the fo1lowing way: Administrative personnel are assumed to work the same number of hours as production workers, and the number of hours in these two categories have been added for the cement industry. The same assumption is made for employees in limestone quarries. Employment in limestone quarries has been obtained by assuming that the proportion of 1imestone quarry emp10yees out of total quarry employment has remained at the 1973 leve1 throughout. This was the only year for which separate data for 1imestone quarries were available.
Sources: See next ~age.
15.
Thus, considering the labor productivity difference, the Swedish \'Jage
cost per ton of cement surpassed the U.S. wage cost in 1971 and was as 51 % higher than the American wage cost in 1974. (See also figures 7
and 8 below.) ,
Energy Consumption
At the same time as labor productivity has increased in both countries, fuel consumption has also been reduced, as illustrated in figure ~. The reduction has been about 25 % in the United States and 20 % in Sweden, but the remaining difference is still very large. For comparison, the fuel consumption in West Germany during the same period is also shown in figure 6 and ;s found to be still lower than that in Sweden.
1".2. PrQduct;on Costs and Cement Prices - ~ ~ ~.~ ~ ~ ~- ~. ~ - ~ ? ~ - ~-
ErQd~c!iQn_CQs!s_a~d_C~m~n! Eric~~
In view of the fact that there have been 50-60 cement companies in the
United States throughout the period and only one or two in Sweden, one might expect the pressure of competition to have keptthe price lower in the Un i ted Sta tes than i n SVJeden. A look a t fi gures (7 and 8, hmvever, wi 11 show that just the opposite has been true. The price of cement has been
Be l ol1~_ to F"igure 5 preced i ng page: S:lUrces: ~a~oE Er~d~e!:i~i!y.:.
ID S I ndustri for each year .• FEA-PCA Proceedings, op. cit., pp. 25-27.
~age_r~t~ in_m~n~f~c!:uEi~g Swedish Employers~ Confederation, Direct and Total Wage Costs for Workers, Various issues. U.S. figures for 1950 and 1955 have been obtained by chaining together Mith data from Council of Econoroic Advisers, Economic Report of the President, Janua~y 1966 (Washington: USGPO, 1966), p. 243. Swedish hourly salarles 1950-1973 have been obtained trom SOS L6ner 1973, Part 2, and for 1974-1975 from Allmän Månadsstatistik. Total wage costs have been obtained by adding fees and charges for social benefits according to information from the Swedish Employers.' Confederation. The wage rate is expressed in current prices. The Swedi~h figures have beenfconverted into dollars using the average offlcial exchange rate or each year. -
Figure 6 Fue1 consumption in cement production in the United States, Sweden and West Germany 1950-74 Million BTU per short ton of cement
Million BTU per short ton of cement
KVJhjmetri c ton
1900
1600 j
L j
1000
T
74-----4---------~~~~------------_r------------~
6 4-----4---------------r------------+~~;_--------~
5 4-----~------~------r-------------_T---------------
\ \ \ \
Sweden ,/
4 ~----+_------------_+--~,----------r---~~------~ \
" \ \ \
/ ""'""''-",,\
W Germany \ /\
I \ --'
3 4-----4---------------~------------~------------~
1950 1960 1970 1980 Year
16.
Sources: G. A. Schroth, II Grade Preheater Ki 1 n Systems II i n FEA-PCA Proceedi ngs, op. cit., p. 253. Cementa AB.
Figure·7 Production Costs and Cement Prices in the United States 1950-1974
S/ton
30
25
20
15
10
5
1950 55
USA
Capita 1,
profit, etc.
Fue1
Labor
60 65
Sources: See next page.
Price
70 75 Year
17.
Figure 8 Production Costs and Cement Prices in Sweden 1950-1975
S/ton
30
25
20
15
10
5
_ 1950 55
Sweden
Capita 1 ,
profit, etc
60 65
Price
70 75 Yea
Fi gure 7
Sourees: Cement Priee:
Eleetrieity Cost:
Fue1 cost:
Labor eost:
Fi gure 8
Sources: Cement price:
Electricity cost:
Fuel eost:
Labor eost:
HL
1950-70: FEA-PCA Preceedings ... , ep. eit.,
p. 43. 1971-74: U.S. Bureau of Mines, Minerals Yearbook 1974, Vol. l, p. 283.
E1eetrieity eonsumption: G.A.Schroth,
op. e it ., p. 236. Electricity price: Edisen Electric Institute, Histerical Statistics of the E1ectrie Uti1ity Industry, EEl Publieation 62-69,
New York, 1962, table 45. EE1, Statistical Yearbook of the Electic Utility lndustry for 1975, EEl Pub1icatien No. 76-51, New Yerk, 1976, table 60 S.
Total energy use: PCA, Conservation Potential .. op. eit. p. 15. Distributien of energy consumption on fue1 type: FEA-PCA Proceedings ... , op. cit., p. 35. Price of eeal: Minerals Yearbeok, various issues. Price of gas: American Gas Association, Gas Faets, 1950, 1951, 1975, 1976, Arlington, V Price of oil: Platt-s Oil Pr;ce Handbook and Oi 1 manac 1976, New Yerk, r~c Graw-Hi 11 lnc, 1976
FEA-PCA Proceedings ... , ep. cit., pp. 25-27.
SOS Industri, National Central Bureau of Statistics, Stockholm, various issues.
Electricity consumptien: Ibid. Electricity price: State Power Board.
Fuel consumptien: SOS Industri Fuel priees: Svenska Petreelum Institutet,
En bok om olja, Stockhelm, SPI, 1971; Svenska Esso AB, Oljeåret 1975; SOS Utrikeshandel, various issues. Fi gure 5 i n the present paper.
19 .
. 13 to 63 % higher in the U.S. than in Sweden, the priee differenee being
especially great around 1960.
eost differenees seem to explain only part of this difference. As shown
in figures 7 and 8, the total variable cost (labor plus fuel and e1ectrici
ty) was higner in the U.S. until 1965 but has since been lower. The U.S.
labor cost per ton of cement was substantially higher than the corresponding Swedish figures during the 1950~s, approximate1y the same during the 1960~s and early 1970~s and the n 20 % 10wer in the last few years due to
extremely rapid Swedish wage increases, coupled with deva1uation of the dollar. Swedish fuel costs per ton of cement were considerab1y higher than
those in the United States in the 19.50 1.s., onTy s.lightly higher in the 1960's, rising again in the 1970~s in relation to the U.S. fue1 eosts. Thus, even
though the U.S. fuel consumption was about 40 % higher than the Swedish one throughout the period, the fuel costs were lower than in Sweden,
primarily due to the avai1abi1ity of cheap domestic natural gas and coal.
Sweden, lacking both of these resources, had to import fueT and eame to rely primarily on oil.
However, the avai1ability in Sweden of cheap hydro power led to low e1ectricity prices which show up in Dur calculation. Thus, the cost of electricity
per ton of cement was only 1/3 of the U.S. electricity eost in 1950. In absolute terms, the cost differenee was about the same throughout the period. Taken together, fuel and electricity costs have been roughly the
same in both countries until 1971 when fuel costs began to rise in Sweden.
The overall conclusion one can draw from this price and cost comparison is that gross profit per ton of cement has been very substantially higher in the United States than in Sweden during the entire period. It has grown from $ 7.18 per ton in 1950 to $ 16.52 in 1974, while the corresponding
Swedish figures are $ 5.50 and $ 9.77. Even if capital costs in the U.S . . had been higher than in Sweden, which may have been the case but is re-
lative1y un1ikely, it seems fair to conelude that net profits must have been considerably higher per ton in the U.S. than in Sweden over the
whole 24-year period. It is apparent, however, that the overcapacity which existed in the U.S. cement market in the 1960~s put a substantial
downward pressure on prices and thereby profits. Given the general rate
of inflation in the economy, the profits squeeze may have been serious in many companies by the early 1970~s - but worst in Sweden where the general rate of inflation has been higher than in the United States.
20,.
In order to put these results in same perspective, it can be mentioned that the Portland Cement Association has estimated that the investment cost of a new cement plant in the U.S. was $ 88 per metric ton in 1974.
1)
Assuming 20 years- depreciation and 15 % discount rate, the capital cost
amounts to $ 14 per ton in 1974 prices. This is on1y slightly less than
the average 1974 gross profit per ton calculated above for the U.S. and over 40 % higher than the ca1cu1ated Swedish gross profit. Although the
development of investment costs per ton of ce~ent over the last 25 years
is not knovJn, it is not likely that investr.1ent in the cer.1ent industry has been very profitable since 1960.2}
v. Het vs. Dry Pl ants - A Theoreti ca l 'Cost Compad son "
. ,
It was argued earlier that all the major eost components are theoretically
lower for preheater dry than for wet process kilns: the investment cost per ton of capacity is lower, and the labor and fuel costs per ton of output are a1so lower. But if this is true, a1so in practiee, how is it possible
that U.S. firms kept investing in wet process kilns until 1975 and that the
wet process share of total U.S. cement production increased unti1 at least 1970?3) How bi g are the cost di fferences between preheater dry and 't,et
process kilns?
In order to gain some idea of the answer to this question, let us make a standardized cost calcu1ation for a typical wet process and dry process installation in 1970 and then a similar camparison for 1975 (af ter the
energy price changes}, using aggregate national price data for both years.'
We will then report on the range of variation in actual costs and input
requirements among individual plants obtained from interviews with cement
firms in both Sweden and the United States.
In table 2 a eompari$on is made of the total cost of production in a new
wet plant in the U.S. and S\veden to that of a new preheater dry plant, us;ng average prices for both countries in 1970 and representative input
requirements. The price assumptions are based on available national price averages for energy and 1abo~ in the stone, c1ay and glass products industry. Theinvestment,cos't per annual ton of plant capacity has been obtained
... ,.'
from a German study. See figure 4.
l) Energy Conservation Potential 't 19 .......•.• , op. C 1. " p. •
2) It is an interesting question for further research what the reasons are for the low profitability in Sweden and whether this is a general phenomenon.
3) Ibid ........... , p. 12.
Tab 1 e 2 Hypothetical Cost Comparison between Dry and Wet Process Cement Plants in the U.S. and Sweden, 1970
Wet method. 600 000 tonslyear Dry method, 600 000 tons/year Requirement per eost, $TfOrl Requirement per eost, t/ton
Cost item Price per unit, $ ton of cement of cement ton of cement of cement U.S. Sweden U.S. Sweden U.S. Sweden U.S. Sweden U.S. Sweden
Coal 0.40 0.68 2.1 ~1BTU 0.0 0.84 0.0 1.40 tlBTU - 0.56
Natural gas 0.38 2.6 MBTU .0.0 0.99 0.0 1. 75 t:lBTU - 0.67
Fue 1 oil 0.49 0.60 0.5 . ~1BTU 04.8 0.25 2.88 0.35 f1BTU 3.1 0.17 1.86
Total fuel 0.40 0.60 5.2 1'1BTU 4.8 2.08 2.88 3.50 HBTU 3.1 1.40 1.86
Eleetrie power 9.50 7.30 0.13 t~Wh 0.10 1.24 0.73 0.14 ~1Wh 0.10 1.33 0.73
Total energy 3.32 3.61 2.73 2.59
Other variable costs 1.00 1.00 1.50 $ 1.50 1. 50 1. 50 1.00 $ 1.00 1.50 1.50
Total variable eosts 4.82 5.11 4.23 4.09
Labor 4.25 3.00 0.45 hours 0.54 l .91 1.62 0.45 hours 0.54 1.91 1.62
Capita 1 1.0 1.00 5.51 $ 5.51 5.51 S.51 4.71 $ 4.71 4.71 4.71
Total production eos t 12.24 12.24 10.85 10.42
Cement price 19.43 13.68 19.46 13.68
Nate: MBTU = ~·1i 11 i on British Thermal Units. 1 MBTU = 293 kWh. Sources: See text. N
22. The investment cost assumptions made for 1970 in table 2 are $ 34.50 per
ton of annual capacity for the wet plant and $ 29.50 for the dry plant. With a 15 % discount rate and 20 years~ depreciation this amounts to a capital cost per ton produced of $ 5.51 and $ 4.71, respectively.
As far as labor requirements are concerned, it is assumed that both plants wou1d require 150 employees in the U.S. and 180 in Sweden, with
an average of 1800 hours worked per year.
The energy consumption (both fuel and electricity) is assumed to be that of best practice plants of the respective kind in both countries. As indicated in the table, the American energy consumption figures are somewhat higher than the Swedish ones in view of the existing differences in
operating practices and product specifications. The distribution on types of fuel corresponds to the averages for the cement industry in each country in 1970.
In spite of the large differences in both prices and input requirements, the overall cost picture turns out to be remarkably similar in the two countries both for the total costs and for the major cost components. The wet method turns out to be about 15 % (or about $ 1.50) more expensive per ton produced than the dry process in both countries. But in Sweden the existing price of cement permitted a profit of only $ 1.50 per ton with the wet method, while the profit margin was $ 3 per ton with the dry method. Due to the considerably higher prices in the U.S., both methods were highly profitable, although the profit margin was about $ 1.50 per ton larger for the dry process.
In table 3, the same comparison-is made using 1975 prices and input requirements. Relative prices have changed considerably, and the distribution on fuel types has changed in line with present trends. Thus, both fuel prices and investment costs have approximately trebled, while the wage rate increased by "only" 140 % in Sweden and by 50 % in the U.S. In this manner the costs of cement production more than coubled in both countries. The dry method is still considerably cheaper than the wet process, hut the absolute cost difference has trebled. At the same time the cement price development has been such that it is no longer possible to cover the costs of production in newly built wet plants even in the United States. On the other hand, the dry method does not seem very profitable either. But this is probably due 1arge1y to the excess supp1y
of cement in the world market during the last several years.
Tab l e 3 Hypothetical Cost Comparison between Dry and Wet Process Cement Plants in the U.S. and Sweden, l ~72-
Het method. 600 000 tons/year 'Requirement per eos t, $/ton
Cost item Price per unit, $ ton of cement of cement . U.S. Sweden U.S. Sweden U.S. Sweden
Coal 1. 12 1..71 4.05 t·1BTU 2.40 4.54 4.10
Natural gas 0.99 0.73 MBTU 0.72
Fuel oil 1. 93 2.09 0.42 t,1BTU 2.40 0.81 5.02
Total fuel 1.17 5.20 MBTU 4.80 6.07 9.12
Electric power 19.20 11 .80 0.13 MBTU O. 10 2.50 1.18
Total energy 8.57 10.30
Other variable costs 1.00 1.00 1.50 $ 1.50 1.50 1. 50
Total variable costs 10.07 11 .80
Labor 6.50 7.20 0.45 hours 0.54 2.93 3.89
Capita l 1.00 1.00 15.60 $ 15.60 15.60 . 15.60
Total production 28.60 31 .29 cost
Cement price 26.52 25.40
Note: MBTU = t·1ill ion British Thermal Units. l MBTU= 293 kWh.
Sources: See text.
Dry method, 600 000 tons/year Requirement per eost, S/ton ton of cement of cement U.S. Sweden U.S. Swed:::n
2.73 MBTU 1.55 3.06 2.65
0.49 MBTU 0.49 0.28 MBTU 1.55 0.54 3.24
3.50 t~BTU 3. l O 4.09 5.89
0.14 MBTU 0.10 2.69 1.18 6.78 7.07
1.50 $ 1.50 1.50 1.50
8.28 8.57
0.45 hours 0.54 2.93 3.89
14.11 $ 14.11 14.11 14.11
23.32 26.57
26.52 25.40
N W .
24.
VI Actualvs TheoreticalCostDifferences between Wet and Dry Plants
Thus, if we look at national averages, it is easy to see why no wet ki1ns have been built in Sweden since 1967 nor in the United States since 1975. But if our cost calculations are at least roughly representative of the industry, there still remains a good bit to be explained. If firms are rational, and if a higher profit is regarded as more
desirable than a lower profit, then hm', can we explain why wet plants continued to be built for so long in both countries? Perhaps the national averages gloss over differences among this seemingly erratic or irrationa1 behavior? between wet and dry plants are not as great in
plants which would exp1ain
Perhaps the eost differences practice as in theory?
In May-June, 1977, the author of this study carried out a number of interviews with representatives of cement firms in both Sweden and the United States, major equipment manufacturers, a consultant firm, and the industry's branch organization in the United States, the Portland Cement Association. Data were gathered for a large number of plants in both
countries. Emphasis was put on plants bui1t in the late 1960's and mid-1970's -- investment costs, operating and price data, and especial1y the
judgements made in connection VJith major investments and the reasons for building the particular type and size of plant.
Looking first at the empirica1 evidence concerning energy, it is
quite clear that suspension preheater and precalciner systems do offer considerable energy savings in comparlson wltn DOtn wet and conventional (long) dry systems. Converted into east terms by using 1976 U.S. energy priees, the difference in energy consumption between
preheater dry and wet process plants amounts to $ 2.00-2.50 per ton of cement. The savings are greatest in fue1s, whereas at least in U.S. operations the electricity consumption is higher in preheater dry than in wet systems. In both dry and wet systems, the Swedish plants seem to be more energy efficient.
The pri ces qu oted for eoa l i n 1977 ranged from $ O. 84/~1BTU ($ 22 per metri c ton) to $ 1.75/MBTU ($ 46. per ton) in the United States. For national
gas the price range was $ 0.78/MBTU to $ 2.15/MBTU, and for fuel oi1 from
$ 1.95/MBTU ($ 12.l0/barrel) to $ 2.03/MBTU ($ l2.60/barrel).
Combined with the differences in fuel requirements observed above, this implies that the fuel east differenee between a wet and a dry plant could
range from $ 2.50 to $ 16.50 per ~etric ton.
As far as electric power is concerned, the prices quoted ranged from 1.5 i/kUh to about 5 i/kWh in the United States and from 2.5 to 3.5 t/kWh in Sweden.
25.
As far as the empirical evidence on the relative labor savinq is con
cerned, the picture is less. cleara If the data were taken at face value, then would indicate that labor costs may even be substantially lower in suspension preheater systems than in ~Iet ones. However, there are simply too few observations to perrnit any conclusions.
r-------~------~~---------~ ~But in this ease the interview results are un-
ambiguous: there are no differences to speak of, given the seale of the installation. At most, there is a difference of one man per shift more in preheater systems (preheater attendant) than in wet systems. The eost difference would am~unt to only $ 0.10-0.20 per ton of cementl~
Turning to capital costs per ton of capacity, the evidence is much less cleara The data were treated in the following
way. The amount of the investment, as reported by each company, was divided by the (gross) additional capacity, yielding a raw figure on the capital eost per ton of annual capacity. Using information as to what items were included in the investQent, it was estimated how much of the total investment for a standard plant2) given in table8 was included, and the raw capital east was adjusted accordingly. Then the adjusted figure was deflated or inflated by a price index to obtain 1974 prices. Unfortunately, no index of investment costs in the cement industry is available, so the United States Wholesale Price Index for industrial commodities was used. The fact that the estimated capital eosts for late-year investments are found to be on the high side is probably an indieation that some better priee index must be found.
But even apart from the priee index problem, it is difficult to make mueh sense of the data. It does not seem possible to say that one type of kiln has consistently higher or lower capital costs than another, nor is it clear even that capital costs decrease with scale. If anything, wet process kilns seem to have lower investment costs per ton than preheater systems. Investment eosts for precalciner systems seem to increase rather than deerease with scale, and the spread in investment costs for SP systems completed in 1976 is between $ 52 and $ 95 per short ton.
l) Assuming three 8-hour shifts 330 days a year with a wage of $ 7.00jhour and an annual production capacity of .5 to l million tons.
2) See the estimate made by the port land Cement Association in Energy Conservation Potential ... , op.sit., p. 19.
26.
What conclusion can be drawn from these rather discouraging results concerning investment costs? Admittedly, the data are very crude, but it appears likely that no adjustment to standardize the data would be sufficierlt to obtain any observable pattern. There are apparently such large differences among plants that it is difficult to speak of a "standardized plant".
There are several reasons why investment costs vary widely among plants. Even though the standardized investment east data referred to must be interpreted with great care, they at least indieate that the cost of installation is higher than the east of the equipment. The installation involves various types of construetion jobs - supports for the kiln, bui1dings and roads, etc. - the cost of which depends on 10ca1 conditions (ski11 and efficiency of 10ca1 contractors, ground conditions, etc.). In addition, the cost of the equipment varies substantially from one case to another. There are only a handful of cement equipment manufacturers in the world (one Danish, a few West German and Japanese; and two American companies which operate mainly on licenses from the other manufacturers) who com-pete in designing and sel1ing whole systems. In order to obtain reference plants they are sometimes wi1ling to offer extremely low prices coupled with substantial guarantees. And of course, prices are always set in negotiations between the cement firm and the equipment manufacturers.
The interview results indicate that opinions in the industry vary widely on whether wet or dry systems have lower investment costs. But it is clear that such statements usually reflect guesses rather than facts; Among all the 14 interviews with cement firms in both the United States and Sweden concerning kilns or plants built in the last 10 years there was on1y one case in which a detailed comparison had been made of what a wet as opposed to a preheater dry installation wou1d cost. In that particular case, the cost comparison came out 20 % lower for the suspension preheater system. But the investment covered only a capacity expansion, not an entire plant. If a whole plant had been considered, the relative east difference probab1y would have been about ha1f as large. In none of the interviews were capital east considerations given 3S the main reason for ehoosing a particular process, and in no ease was the investment eost differenee between the eho sen process and an alternative one deemed to be larger than 15 %.
27.
This is not to say~that investment east differences are unimportant -
af ter all, even a 15 % saving on capJtal east would amount to over $ 2 per ton of cement (i.e. about as much as the energy east differential), if the previausly calculated $ 14 per ton is a representative capital east. But it is clear both that no careful camparison of investment costs was usually made and that fuel saving arguments were given in favor of preheater systems and raw material conditions in favor of wet systems.
To the extent that it is possible to draw any conclusion from this discussion at all, it would seem to be the following. Labor requirements play no role at all in ehoosing among the available technologies. Labor saving arises through increases in scale, regardless of which process is chosen. Even if it is true that capital cost considerations have not played any major role in choosing between alternative technologies in the United States, it is a1so true that U.S. cement installations in recent years have not been particularly large in comparison with European and Japanese plants. Instead, they have been in the size range where wet process kilns seem to have a camparative, even if not absolute, advantage. It is possible, therefore, that as plant and kiln scale continues to increase, capital cost considerations will be come more important - and labor cost differences as wel1. But up until now, energy savings seem to have provided the main argument for the preheater technology.
VII. Reasons for the Delayed Introduction of Suspension Preheater Kilns
The previous discussion has indicated that the only argument for the suspension preheater technology which holds up under scrutiny is the fuel saving argument. Therefore, in order to justify continued investments in wet process plants, one- would have to argue that the fuel saving argument was not applicable to the particular installation considered. There seem to be essentially four reasons why the fuel saving argument may not have been applicable in individual cases.
First at all, one factor which naturally affects the choice between wet and dry process is the moisture content of the raw materials. In our sample of plants, the moisture content varies from 1 % to over 20 %.
The water content of the feed must be reduced to close to zero in any dryoperation. In conventional raw grinding mills (so-called ball mills} there is enough heat generated in the grinding process, although no heat is added, to dry materials containing up to 7 % water. l ) Therefore,
1) U.S. Bureau of Mines, Minerals Yearbook, 1974, p. 298.
28.
there seems to have been a long-standing rule of thumb in the U.S. cement industry that any material with higher than 7 % moisture content
is unsuitable for the dry process.
However, a new type of grinding mill, so-called roller mills, was developed in West Germany, apparently in the early 1960-s. This type of raw mill is used widely in Europe but was introduced in the United States only in 1973. 1) Roller mills use 5 to 15 % less electricity than ball mills, but they are also more amenable to combined grinding-drying than ball mills. By utilizing low-level heat in waste gases from the kiln or preheater it is possible to dry raw materials containing up to 15 % moisture. 2)
By installing additional heating equipment it is possible to dry raw materials with up to 18 % moisture content. The roller mill seems to have been developed precisely to increase the range of utilization of suspension preheater kilns.
At the present time it is not clear whether roller mills per se, require higher or lower investment costs than ball mills. But since they can grind feed of much larger size than ball mills, they mayeliminate a secondary crusher which is usually required. Also, they operate at a much lower noise level than ball mills (reducing the need for noise abatement equipment}. Thus, overall it would appear that the capital cost of roller mills is probably lower than that of ball mil1s. The cost of equipment wear is reported to be about 60 % 10wer than for ball mills. 3}
The imp1ications of this are that in cases where the moisture content exceeds 15 % there may have been reasons to choose the wet rather than the dry process. Even though it seems difficu1t to argue that the raw
materials are wetter, on the averag~, in the United States than in, say, Germany or Sweden, high overland transportation costs and absence of inland water transport facilities may have led to explitation of wet
materials which might not have been used at all in Europe. In the Swedish case, the geography has permitted all but one plant to be located near water, as was noted earlier.
l)U.S. Bureau of Mines, Minerals Yearbook, 1974, p. 298.
2)Gordian Associates, ~R.cit., p. 14. 3)U.S. Bureau of Mines, Minerals Yearbook, 1974, p. 298 .
29.
Another problem which affects the choice between wet and conventional
dry systems on one hand and suspension preheaters and precalciners on
the other is the presence of certain substances in the raw materials
which cause operational difficulties or affect the quality of the product
negatively. The most important of these substances are alkalis (natrium and potassium). If cement containing alkalis is mixed with certain aggreg
ates - prevalent in the Southeastern United States but also in certain other areas, a chemical reaction occurs which causes the concrete to crack.
Therefore, the alkali content is regulated by law. The limit set in the
United States is 0.6 %. However, even customers in areas without reactive
aggregates of ten specify low alkali cement. Other countries also have restrictions on alkali content, although not as stringent. Efforts are currently being made in the United States to enforce the restrictions
only when necessary.
But the presence of alkalis also creates problems in the manufacturing
process itself. Since these are highly volatile substances, they will
simply be blown out with the kiln exhaust gases in open systems such as
wet or conventional dry kilns. But in suspension preheater or precalciner systems which are much more enclosed, alkali content builds up in the
circulating air. If the alkali content of the raw material is low, or if there is just the right amount of sulfur in the raw material or fuel
to balance the alkalis, there is no operational problem in the preheater.
But if there is too much or too little sulfur, the preheater gets plugged
up with sticky material which causes stoppages unless removed.
In order to prevent alkali buildup in suspension preheaters, a so-called
by~pa~shas been developed which allows hot air with high concentrations of alkalis simply to escape from the system. This involves an additional
investment cost and the loss of both energy and raw materials escaping
with the hot air.
It is suggested by some sources l ) that at least some precalcining systems are capable of yielding low-alkali cement with difficult raw materials even
with little or no by-pass. However, this is an issue which needs further investigation.
l) See e.g. Gordian Associates, gp. cit., p. 25.
Given that high alkali content and presence of reactive substances do present difficulties in certain parts of the Unites States and
30.
much less ;n Sweden, the implication is that the alkali problem explains at least some of the observed differences between the two countries in the attitude to the dry process.
The obvious question that arises is whether the alkali problems are unique to the United States or why these difficulties do not seem to have played the same role in other countries. But while it is true that the restrictions on alkali content are more stringent in the U.S. than elsewhere, it is difficult to believe that something as common in the crust of the earth as limestone could vary so much in quality or composition as to be unsuitable for a particular process on one continent but not on another. The following might be at least a partial explanation. Coal is the main fuel used in the cement industry in the United States, while in the 1950-s and 1960-s most European producers switched to oil. Due to the refining process, the sulfur content of fuel oil is held within very narrow margins, even for high-sulfur oils, which means that it is relatively easy to maintain a certain balance between sulfur and alkali in the cement manufacturing process. Coal, on the other hand, usually contains much more sulfur, but above all, the variability of sulfur content is much greater.') This factor, in conjunction with the alkali restrictions in the U.S., provides a third reason for the relatively slow diffusion of suspension preheaters in the United States.
A fourth reason for the delay in introducing the suspension preheater technology, particularly in the United States, is the bad experiences
that several companies had in their early efforts to introduce the new technology. In 1953, just three years af ter the first SP system was installed in Germany, the first preheater system was built in the United
States. This was the fourth such system built in the world until then, which shows that U.S. producers were quick to respond to the new technology. The first SP kiln was followed in the next few years by twelve more. But the majority of these preheaters ran inta several operational difficul ties having to do with a lack of understanding of the sulfur-alkali balance and similar problems. Consequently, many of the se preheaters of ten clogged up, causing considerable downtime and thereby raising both capital and labor costs. About half of the thirteen original U.S. in-
1) Garrett, QR.cit., pp. 273-277.
31.
stallations have now been shut down (some having been replaced by wet kilns!), and between 1955 and 1970 there were only two suspension preheater kilns sold in the United States, one of which has since been shut down. l )
Ironically, therefore, part of the overeapacity in the 1960~s was due to the installation of suspension preheaters, many of which did not function well. 80th the overeapacity and the malfunctioning held back further investment in SP systems. And because of the operational difficulties, the belief became widespread that suspension preheaters were unsuitable to U.S. raw materials and operating requirements.
VIII. Conclusions
The conclusion from the comparison of industrial energy consumption is that international differences in energy prices seem to explain a large part of the differences in energy use as regards both the level and the composition in terms of types of energy.
As far as the examination of the cement industry is concerned, the study started out with the notaion that a comparison between the
United States in Sweden in the choice of cement production technology would be a simple illustration of substitution between energy and other factors of production in view of the differences in relative factor prices, especially relative energy prices. It was soon discovered, however, that the suspension preheater process can be regarded as theoretically superi or to the wet process in almost all respects. The problem th~n be
came that of explaining why the rate of diffusion of the new process has differed among countries, particularly between Sweden and the United States. It was shown in a cost comparison of the wet and the dry process, based
on national average data, that differences in relative factor prices must have been a major influence, and that the drastic price changes which took place between 1970 and 1975 probably constitute the major
reason why investments in the wet technology have dwindled to zero.
l) Garrett, QQ. cit., pp. 273-277.
32.
However, it has also been shown that the re are many factors which in actual
practice reduce the theoretical cost differences quite drastically. The range of variation among plants in prices, raw materials, the market situa
tion etc., is very large indeed. In addition, despite efforts to standardize
for differences among plants in type and size of kiln, year of installation,
etc., it proved very difficult to find any sensible patterns in the data
other than with respect to energy.
A final word about the future might be in order. Given the enormous spread between best practice and average practice plants in the United
States - much larger than in Sweden e.g. - it appears highly probable that rising energy prices will lead to drastic structural changes within the industry. This process has already been analyzed for Sweden. l ) Interest
ing questions arise as to whether the U.S. cement industry will be able
to effect the necessary changes, given the long history of experience with
wet plants, the existing market structure, and the low profitability in recent years. There has been a number of cases recently of European
firms buying up old U.S. plants in order to acquire market shares, then
replacing them with new, larger equipment. This process is likely to continue unless prevented through government policy and is likely to
yield a higher degree of both efficiency and concentration of ownership.
l) See B. Carlsson, II Industrins energiförbtukning 1974-80" (Industrial Energy Consumption 1974-80 11
), appendix 7 in Industriens Utredningsinstitut,
IUI:s långtidsbedömning 1976. Bilagor (IUI~s Long Term Survey 1976. Appendix Volume) (101, Stockholm, 1976), pp. 277-87.
WORKING PAPERS (Missing numbers indicate publication elsewhere)
1976
l. Corporate and Personal Taxation and the Growing Firm by Ulf Jakobsson
7. A Micro Macro Interactive Simulation Model of the Swedish Economy. Preliminary model specification by Gunnar Eliasson in collaboration with Gösta Olavi
8. Estimation and Analysis with a WDI Production Function by Göran Eriksson, Ulf Jakobsson and Leif Jansson
1977
Il. A Comparative. Study of Complete Systems of Demand Functions by N Anders Klevmarken
12. The Linear Expenditure System and Demand for Housing under Rent Control by Per H ögberg and N Anders Klevmarken
14. Rates of Depreciation of Human Capital Due to Nonuse by Siv Gustafsson
15. Pay Differentials between Government and Private Sector Employees in Sweden by Siv Gustafsson
1979
20. A Putty-Clay Model of Demand Uncertainty and Investment by James W. Albrecht and Albert G. Hart
1980
25. On Unexplained Price Differences by Bo Axell
26. The West European Steel Industry - Structure and Competitiveness in Historical Perspective by Bo Carlsson
27. Crises, Inflation and Relative Prices in Sweden 1913-1977 by M ärtha Josefsson and Johan Örtengren
-·2 -
33. The Demand for Energy in Swedish Manufacturing by Joyce M. Dargay
34. Imperfect Information Equilibrium, Existence, Configuration and Stability by Bo Axell
1981
35. Value Added Tax: Experience in Sweden by Göran Normann
36. Energi, stabilitet och tillväxt i svensk ekonomi (Energy, Stability and Growth in the Swedish Economy) by Bengt-Christer Ysander
37. Picking Winners or Bailing out Losers? A study of the Swedish state holding company and its role in the new Swedish industrial policy by Gunnar Eliasson and Bengt-Christer Ysander
38. Utility in Local Government Budgeting by Bengt-Christer Ysander
40. Wage Earners Funds and Rational Expectations by Bo Axell
41. AVintage Model for the Swedish Iron and Steel Industry by Leif Jansson
42. The Structure of the Isac Model by Leif Jansson, Tomas Nordström and Bengt-Christer Ysander
43. An Econometric Model of Local Government and Budgeting by Bengt-Christer Ysander
44. Local Authorities, Economic Stability and the Efficiency of F iscal Policy by Tomas Nordström and Bengt-Christer Ysander
45. Growth, Exit and Entry of Firms by Göran Eriksson
47. Oil Prices and Economic Stability. The Macroeconomic Impact of OH Price Shocks on the Swedish Economy by Bengt-Christer Y sander
48. An Examination of the Impact of Changes in the Prices of Fuels and Primary Metals on Nordic Countries Using a World Econometric Model by K. S. Sarma
- 3 -
50. Flexibility in Budget Policy. Changing Problems and Requirements of Public Budgeting by A. Robinson and B.-C. Ysander
51. On Price Elasticities in Foreign Trade by Eva Christina Horwitz
52. Swedish Export Performance 1963-1979. A Constant Market Shares Analysis by Eva Christina Horwitz
53. Overshooting and Asymmetries in the Transmission of Foreign Pr ice Shocks to the Swedish Economy by Hans Genberg
54. Public Budgets in Sweden. A Brief Account of Budget Structure and Budgeting Procedure by Bengt-Chris~er Ysander
55. Arbetsmarknad och strukturomvandling i de nordiska länderna av Bertil Holmlund
56. Central Control of the Local Government Sector in Sweden by Richard Murray
58. Industrial Subsidies in Sweden: Macro-economic Effects and an International Comparison by Bo Car lsson
59. Longitudinal Lessons from the Panel Study of Income Dynamics by Greg J. Duncan and James N. Morgan
1982
60. Stabilization and Growth Policy with Uncertain OU Prices: Some Ru1es of Thumb by Mark Sharefkin
61. Var står den nationalekonomiska centralteorin idag? av Bo Axell
63. General Search Market Equilibrium by James W. Albrecht and Bo Axell
64. The Structure and Working of the Isac Model by Leif Jansson, Thomas Nordström and Bengt-Christer Ysander
-·4 -
65. Comparative Advantage and Development Policy Twenty Years Later by Anne O. Krueger
67. Computable Multi-Country Models of Production and Trade by James M. Henderson
68. Payroll Taxes and Wage Inflation: The Swedish Experiences by Bertil HolmIund (Revised, September 1982).
69. Relative Competitiveness of Foreign Subsidiary Operations of a Multinational Company 1962-77 by Anders Grufman
70. Optimization under nonlinear constraints by Leif Jansson and Erik Mellander
71. Technology, Pricing and Investment in Telecomnnunications by Tomas Pousette
72. The Micro Initialization of MOSES by James W Albrecht and Thomas Lindberg
73. Measuring the Duration of Unemployment: A Nate by Anders Bj örklund
74. On the Optimal Rate of Structural Adjustment by Gunnar Eliasson
75. The MOSES Manual by Fredrik Bergholm
76. Differential patterns of Unemployment in Sweden by Linda Leighton and Siv Gustafsson
77. Household Market and a Nonmarket Activities (HUS) - A Pilot Study by Anders Klevmarken
78. Arbetslöshetsersättningen i Sverige - motiv, regler och effekter av Anders Björklund och. Bertil HolmIund
79. Energy Prices, Industrial Structure and Choke of Technology; An International Comparison with Special Emphasis on the Cement Industry by Bo Carlsson